A computer system that processes a large amount of data manages data by using a large capacity storage system (for example, a cloud storage) provided separately from a host computer. One of use cases of the cloud storage is a case where a storage disposed in a data center (hereinafter, referred to as on-premises) of a client controls storage of data in the cloud storage. In this case, the host computer is unaware of the cloud storage.
Furthermore, an operation may be executed by a computer and a virtual machine on a cloud, by using data stored in the cloud. Thus, an operation using a large amount of resources at a time, such as data analysis processing, can be implemented at a low cost.
PTL 1 discloses a technique of storing a copy of data, stored in a storage in the on-premises, in the cloud. Differential data between the on-premises and the cloud is periodically stored in the cloud. In the technique, compression, encryption, and the like are performed, on the on-premises side, on data to be stored in the cloud, and a data structure is recognizable by the storage in the on-premises. Furthermore, data can be stored in a plurality of clouds.
A device virtualization function (external storage connection function) has been known in which a storage device is virtualized in a storage. The device virtualization function is a function of mapping a storage device in an external storage in an upper level storage, and providing as data in the upper level storage to the host.